What are GPT-5.5 Instant's benchmark scores?

As of 24 Oct 2026, GPT-5.5 Instant scores 58.0 ±5.4 on the BenchLeader Index, strongest in instruction following (70) and weakest in maths (49). That rests on 17 independent benchmark results, and every score on this page links to its publisher.

How does GPT-5.5 Instant rank?

GPT-5.5 Instant is #163 of 785 ranked configurations on 24 Oct 2026. 282 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.

How much does GPT-5.5 Instant cost per 1M tokens?

As of 24 Oct 2026, GPT-5.5 Instant costs $5.00 per 1M input tokens and $30.00 per 1M output tokens, $11.25 blended (three input tokens per output token).

When was GPT-5.5 Instant released?

OpenAI released GPT-5.5 Instant on 5 May 2026.

OpenAIReasoning modelCompare with another model

GPT-5.5 Instant

GPT-5.5 Instant is an OpenAI proprietary reasoning model, released 5 May 2026. Its best configuration ranks #163 of 785 on the BenchLeader Index at 58.0 ±5.4, in the upper half. The ± is the point: 282 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in instruction following (70) and lowest in maths (49). At $11.25 per million tokens blended it is among the most expensive ranked models. Output speed of 156 tokens per second puts it in the fastest quarter, with a first answer in 13.7 s. Last measured 22 Oct 2026.

Blended price: $11.25/M; $5.00 in · $30.00 out
Output speed: 156 tok/s; measured by Artificial Analysis
First answer: 14 s
Context: 400k
Full answer: 17 s; median, reasoning included
Cost per run: $0.691; one full Intelligence Index run
Released: 5 May 2026

How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.

Overall index58
Reasoning55
Coding61
Agents & tools50
Maths49
Knowledge67
Instruction following70
Human preference66
Multimodal58
Long context60
Composite59

Versions

OpenAI has shipped 2 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.

Model	Released	Index	Rank
GPT-5.5 Instantthis page	5 May 2026	58.0	#163
GPT 5.1 Instant	–	–	–

How its index has moved

8 Sept 2026 to 21 Oct 2026

The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.

Benchmark results

See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

Benchmark	not stated	Source
GPQA Diamond	82.5%#112	Epoch AI Benchmarking Hub
Arena Hard Prompts	1491#46	Arena (LMArena)
GPQA Diamond (AA)not in index	84.7%#135	Artificial Analysis
Humanity's Last Exam (AA)not in index	21.6%#206	Artificial Analysis
CritPt	2.6%#189	Artificial Analysis

Coding

Benchmark	not stated	Source
SciCode	52.5%#85	SciCode
Arena Coding	1513#49	Arena (LMArena)
SciCode (AA)not in index	52.5%#85	Artificial Analysis
Terminal-Bench 4.0 (AA)not in index	12.6%#79	Artificial Analysis
Terminal-Bench 2.1 (AA)not in index	34.8%#149	Artificial Analysis

Agents & tools

Benchmark	not stated	Source
Terminal-Bench Hard	42.4%#46	Artificial Analysis
τ²-Bench Telecom (AA)not in index	49.4%#198	Artificial Analysis
GDPval-AA v2.1not in index	0.4%#123	Artificial Analysis
τ³-Banking (AA)	11.8%#136	Artificial Analysis
AutomationBenchnot in index	36.4%#105	Artificial Analysis
GDP.pdf (AA)not in index	20.2%#63	Artificial Analysis

Maths

Benchmark	not stated	Source
FrontierMath Tiers 1–3	26.3%#78	Epoch AI Benchmarking Hub
FrontierMath Tier 4	2.4%#68	Epoch AI Benchmarking Hub
OTIS Mock AIME	68.1%#146	Epoch AI Benchmarking Hub
Arena Maths	1460#64	Arena (LMArena)

Knowledge

Benchmark	not stated	Source
AA-Omniscience	11.2#92	Artificial Analysis
AA-Omniscience: accuracynot in index	46.3%#83	Artificial Analysis
AA-Omniscience: non-hallucinationnot in index	34.7%#190	Artificial Analysis

Instruction following

Benchmark	not stated	Source	Trend
IFBench	71.5%#54	Artificial Analysis
Arena Instruction Followingnot in index	1456#56	Arena (LMArena)

Human preference

Benchmark	not stated	Source
Arena Text	1473#43	Arena (LMArena)
Arena Creative Writingnot in index	1455#32	Arena (LMArena)
Arena Multi-turnnot in index	1483#34	Arena (LMArena)
Arena Longer Queriesnot in index	1476#54	Arena (LMArena)

Multimodal

Benchmark	not stated	Source	Trend
Arena Vision	1250#63	Arena (LMArena)

Long context

Benchmark	not stated	Source
AA-LCR	70.0%#215	Artificial Analysis
Arena Documentnot in index	1460#26	Arena (LMArena)
MLCRnot in index	31.7%#13	Artificial Analysis

Composite

Benchmark	not stated	Source	Trend
Epoch Capabilities Indexnot in index	142.4#101	Epoch AI Benchmarking Hub
AA Intelligence Index v4.3.2	26.0#158	Artificial Analysis

What a task costs

Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.

Workload	Tokens in / out	Cost	With caching	Time
Chat reply	400 / 300	$0.011	–	15.7 s
Summarize a 30-page report	12,000 / 600	$0.078	–	17.6 s
Code edit	6,000 / 1,500	$0.075	–	23.4 s
Agentic coding session	60,000 / 4,000	$0.420	–	39.4 s
Structured extraction	2,000 / 200	$0.016	–	15.0 s

New this month

Questions people ask

What are GPT-5.5 Instant's benchmark scores?: As of 24 Oct 2026, GPT-5.5 Instant scores 58.0 ±5.4 on the BenchLeader Index, strongest in instruction following (70) and weakest in maths (49). That rests on 17 independent benchmark results, and every score on this page links to its publisher.
How does GPT-5.5 Instant rank?: GPT-5.5 Instant is #163 of 785 ranked configurations on 24 Oct 2026. 282 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
How much does GPT-5.5 Instant cost per 1M tokens?: As of 24 Oct 2026, GPT-5.5 Instant costs $5.00 per 1M input tokens and $30.00 per 1M output tokens, $11.25 blended (three input tokens per output token).
When was GPT-5.5 Instant released?: OpenAI released GPT-5.5 Instant on 5 May 2026.

Data as of 24 Oct 2026. Compare with another model.

Cite as: BenchLeader, “GPT-5.5 Instant: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/gpt-5-5-instant.